Alzheimer's disease (AD) is a progressive neurodegenerative disorder. Findings from the literature, and preliminary data reported in this application, suggest lipid measures, including brain-derived cholesterol species and lipid peroxidation products, may produce readily detectable signatures in the blood of AD patients. These signatures may be indicators of AD-associated neurodegeneration, producing candidate biomarkers of disease progression. A blood-based biomarker would be superior to CSF-based or brain imaging biomarkers with regard to cost, invasiveness and feasibility. A biomarker of neurodegeneration would have several applications, including use as a surrogate or secondary outcome measure in clinical trials of emerging therapies. Thus far, few longitudinal studies have been conducted in this area. This is important because the rate of change or accumulation of a biomarker may be a better indicator of disease progression than a single value or range at one time point. Such a biomarker could be used as a surrogate or secondary outcome measure in clinical trials of emerging therapies for AD, or may be a prognostic indicator of disease progression. We propose a proof-of-concept study to explore the clinical utility of 24S-OHC, 24S-OHC/27-OHC ratio, 24S- OHC/cholesterol ratio, and F2a-isoprostanes as blood-based biomarkers of neurodegeneration and, therefore, AD progression. We are currently following 75 well-characterized individuals, 25 each with early AD, MCI, and age-matched controls in a longitudinal imaging study. Participants are imaged 4 times over a year, at which time they also provide blood samples and have a thorough clinical and cognitive assessment. Blood samples and morphometric MRI scans collected as part of this study will be used for the present proposal, providing a unique opportunity to identify candidate blood markers of neurodegeneration within the RFA 2-year time frame. Not only will we be able to assess whether these lipids are associated with disease progression but we will be able to correlate the biomarkers to morphometric MRI, an established measure of brain atrophy, to determine whether these blood-based lipids are indeed indirect biomarkers of neurodegeneration. Analyses will: (1) Estimate the variability of plasma levels of each biomarker both within and between individuals at baseline, 3, 6, and 12 months and determine factors that effect this variability; (2) Compare the cross-sectional and longitudinal variations and trajectories in biomarkers for the three groups of individuals with varying AD pathology; (3) Determine whether baseline biomarker levels predict and/or correlate with change in tests of memory and executive functioning; (4) Examine the cross-sectional and longitudinal relationship between the plasma lipid biomarkers and brain atrophy or change in brain atrophy, as measured by morphometric MRI, in regions most affected early by Alzheimer-associated neurodegeneration (e.g. hippocampus, entorhinal cortex) and control regions not affected in early Alzheimer's (e.g. thalamus, putamen). This R21 application proposes a proof-of-concept study to explore the clinical utility of blood-based lipid biomarkers, including 24S-hydroxycholesterol and F2a-isoprostanes, as indicators of Alzheimer's disease (AD)-associated neurodegeneration. A validated blood-based biomarker would be superior to more invasive and costly CSF-based or brain imaging biomarkers. Such a biomarker of neurodegeneration would have several applications, including prognosis after the onset of dementia, prognosis of conversion from mild cognitive impairment to dementia, or use as a surrogate or secondary outcome measure in clinical trials of emerging therapies for neurodegenerative disorders. [unreadable] [unreadable] [unreadable]